Brain Tumor Detection using Machine Learning

Digital Brain tumor detection is a critical challenge in medical imaging and diagnosis, with early detection being vital for effective treatment and management. With the advent of machine learning (ML) techniques, significant progress has been made in automating brain tumor detection from medical images such as MRI scans [1], [2]. This paper presents a comprehensive study on the application of various machine learning algorithms for brain tumor detection, with a focus on the Support Vector Machine (SVM) model. The objective is to evaluate the performance and accuracy of SVM compared to other popular machine learning models, including Decision Trees, Random Forests, K-Nearest Neighbors (KNN), and Logistic Regression. In this study, a dataset of MRI brain images is pre-processed using techniques like normalization and feature extraction. Several classification algorithms are applied to detect and classify brain tumors as benign or malignant. Among all tested models, the SVM outperforms the others in terms of accuracy, precision, recall, and F1-score. The SVM model uses a kernel trick to map input features into higher-dimensional spaces, providing better classification boundaries and generalization capabilities. This enables the SVM model to handle non-linear data more efficiently than linear classifiers. Additionally, SVM's ability to work with a small number of training samples and high-dimensional data further enhances its performance.